LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Meta Llama 3 8B by meta-llama

 »  All LLMs  »  meta-llama  »  Meta Llama 3 8B   URL Share it on

  538126   
  Autotrain compatible   En   Endpoints compatible   Facebook   Llama   Llama-3   Meta   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/meta-llama/Meta-Llama-3-8B 

Meta Llama 3 8B Benchmarks

MMLU Pro: 24.55
GPQA: 7.38
MUSR: 6.24
BBH: 24.5
IFEval: 14.55 vs 88 (so35)-83.5%
ARC: 59.22 vs 96.7 (so35)-38.8%
HellaSwag: 82.02 vs 95.3 (gpt4)-13.9%
MMLU: 66.49 vs 88.3 (so35)-24.7%
TruthfulQA: 43.95 vs 59 (gpt4)-25.5%
WinoGrande: 77.11 vs 87.5 (gpt4)-11.9%
GSM8K: 45.34 vs 96.4 (so35)-53%
MATH Lvl 5: 4.53
LLME Score: 0.46871
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Meta Llama 3 8B (meta-llama/Meta-Llama-3-8B)

Meta Llama 3 8B Parameters and Internals

Model Type 
text generation, multimodal
Use Cases 
Areas:
commercial use, research
Applications:
assistant-like chat, natural language generation
Primary Use Cases:
pretrained models adapted for various NLG tasks
Limitations:
Only tested in English, Not all scenarios addressed.
Considerations:
Developers can fine-tune Llama 3 for languages beyond English, compliance required.
Additional Notes 
Some trade-off between model helpfulness and alignment likely unavoidable.
Supported Languages 
English (Primary)
Training Details 
Data Sources:
publicly available online data
Data Volume:
15 trillion tokens
Methodology:
auto-regressive transformer architecture, supervised fine-tuning, reinforcement learning with human feedback (RLHF)
Context Length:
8000
Training Time:
7.7M GPU hours
Hardware Used:
H100-80GB (TDP of 700W)
Model Architecture:
optimized transformer architecture
Safety Evaluation 
Methodologies:
red teaming, adversarial evaluations
Findings:
residual risks may remain
Risk Categories:
child safety risks, security risks, bias risks
Ethical Considerations:
Openness, inclusivity, and helpfulness valued.
Responsible Ai Considerations 
Fairness:
Consideration given to bias and fairness during development.
Transparency:
Safety evaluations and risk assessments detailed.
Accountability:
Developers advised to perform application-specific safety testing.
Mitigation Strategies:
Red teaming, adversarial evaluations, safety mitigations included.
Input Output 
Input Format:
text
Accepted Modalities:
text
Output Format:
text and code
Release Notes 
Version:
3.1-8B
Date:
2024-04-18
Notes:
Released with improved performance for dialogue use cases.
LLM NameMeta Llama 3 8B
Repository 🤗https://huggingface.co/meta-llama/Meta-Llama-3-8B 
Model Size8b
Required VRAM16.1 GB
Updated2025-04-28
Maintainermeta-llama
Model Typellama
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Quantized Models of the Meta Llama 3 8B

Model
Likes
Downloads
VRAM
Llama 3 8B Bnb 4bit197806795 GB
Meta Llama 3 8B AWQ095975 GB
Monah 8B Uncensored V0.2 Gguf51383 GB
Monah 8B Uncensored Gguf41203 GB
Monah 8B Uncensored V0.2 Gguf51313 GB
Llama3 8B Chips 4bit075 GB
Hyper X Llama 3 Small Python055 GB
Openchat 3.6 8B 20240522 8bit2208 GB
Llama 3 8B GPTQ 4 Bit62585 GB
...at 3.6 8B 20240522 8 0bpw EXL2248 GB

Best Alternatives to Meta Llama 3 8B

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB4241103
UltraLong Thinking4192K / 16.1 GB692
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB128915
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB362540
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
....1 1million Ctx Dark Planet 8B1048K / 32.3 GB262
...a 3 8B Instruct Gradient 1048K1024K / 16.1 GB26621679
A11024K / 16.1 GB8530

Rank the Meta Llama 3 8B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 46763 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer